Decision Analysis to Inform Screening of the Young: An Interactive Newborn Screening Model

Decision analytic models that project the outcomes of population-based newborn screening compared with clinical identification can inform newborn screening policymaking. A flexible decision analytic model that could be adapted for the many and growing number of conditions being considered for newborn screening would help expedite decision making while reducing the time and resources required to develop new condition-specific models. To address this need, we developed Decision Analysis to Inform Screening of the Young (DAISY), an interactive and adaptable generic newborn screening model that incorporates user-supplied inputs. To illustrate the use of this tool, we present a model of newborn screening for metachromatic leukodystrophy (MLD). DAISY projects that newborn screening would identify a greater number of MLD cases and improve survival outcomes over a 5-year time horizon compared with clinical identification. In a US birth cohort of 3.6 million newborns, we project that 36 infants with early-onset MLD cases would be identified per year through newborn screening, each of whom would be treatment eligible. In contrast, in the absence of screening, only 11.5 infants with early-onset MLD cases would be identified, 10 of whom would be treatment eligible. The number of children surviving without motor impairment approximately 5 years after symptom onset is projected to be 32.8 with newborn screening and 9.1 in the absence of screening. DAISY can be used to evaluate other newborn screening candidate conditions through specification of input parameters such as population size, clinical validity of the screening algorithm, and treatment effectiveness.